Table of Contents

Summary

The Hudson River Valley is of primarily erosion-resistant uplands overlain with glacially deposited unconsolidated silts and clays lining most valleys of the tributaries. Sediment sources are largely confined to a veneer of glacial deposits. Sand enters the system from local concentrations of glacial sand bodies. In addition, sand from the oceanic Coastal Plain migrates into the lower estuary under tidal influences. The Hudson River estuary exports an average of 436,000 metric tons of sediment per year to the NY-NJ harbor, although this number can vary by a factor of more than two between years. Roughly 212,000 metric tons of sediment is input on average from the upper Hudson River and Mohawk watersheds to the estuary. Another 224,000 metric tons is input on average from mid-Hudson tributaries draining into the estuary. The spring freshet enters the tidal river with high sediment concentrations; most of that sediment settles and is resuspended multiple times in its trajectory through the tidal river. Geochemical data suggest sediment is initially transported to the lower Hudson (mile 43) in timescales of 0-4 years, but its residence time is on the order of 20 years. Physical transport observations found that sediment fluxes at Poughkeepsie (mile 72) reflected sediment inputs associated with the spring freshet from the Mohawk within 20-25 days. More investigations of the mechanisms and volumes of sediment transport in the tidal freshwater Hudson are needed in order to resolve the contrast in timescales between the physical transport observations and the geochemical data.

Patterns of erosion and deposition are very heterogenous throughout the tidal river, and every reach of the river has both erosional and depositional portions. These observations suggest there is a significant volume of reworking of sediment in addition to throughput of watershed sources. Large portions of the channel have experienced 1-2 m of erosion in the last 100 years. Analysis of historical bathymetric data indicates that a section of the Hudson River channel lost an estimated 3 ,000,000 tons of sediment between ca. 1939 and 2002 (50,000 tons/yr average) by subaqueous erosion, increasing in depth by as much as 4 m in places. These changes are attributed to the influence of land fill and pier construction on both sides of the river, which reduced the cross-sectional area by 10-20% in some places.

Two primary trapping zones are known in the lower estuary, one centered at river km 13 to the south of the George Washington Bridge, and the other near river km 23 to the north of the bridge. The estimated mass of sediment trapped in the lower estuary during studies in 1998 and 1999 was approximately 300,000 metric tons per year. This is slightly less but comparable to the estimated sediment flux being delivered from the Hudson-Mohawk, as discussed above. Spatial patterns of deposition are likely controlled not only by estuarine processes, but also by the accommodation space afforded by historical dredging of the harbor. The dredging activities typically took place in regions of high sediment trapping rates. The increased depth associated with dredging tends to alter the hydrodynamics, favoring sediment trapping. Deposition rates in dredged areas are estimated to average 14 cm/yr, although the deposition rate is not uniform. In the entire estuary, the 1000-year rate of sediment accumulation is about 2 mm/year, close to the rate of sea-level rise. However, tidal accumulation rates are extremely rapid, highly intermittent and subject to erosional episodes, yielding a net seasonal accumulation rate of approximately 10 cm/year.

Literature Review

Geologic Setting

The following description of the geologic setting of the Hudson River Valley is taken from Bokuniewicz (2005). The Hudson and Mohawk Rivers transverse predominantly erosion-resistant uplands, although the valley itself occupies a terrain of erosion-susceptible shale. The lower Hudson crosses six geologic terrains: (1) from Troy to Cornwall, the river runs through a valley in the Appalachian Ridge and Valley Province; (2) from Cornwall-on-Hudson southward to Peekskill, the river cuts through the Hudson Highlands, a band of resistant, Precambrian crystalline rocks; (3) below Peekskill, the west bank of the river skirts the rocks of the Newark Basin; (4) the east bank is formed by high-grade metamorphic rocks of the Manhattan prong of the New England Uplands – Precambrian and Lower Paleozoic shists, marble, quartzite, and gneiss; (5) the Hudson discharges into the Upper and Lower bays on New York Harbor; and (6) spills out across the unconsolidated sediments of the Coastal Plain to the Atlantic Ocean.

The rocks of these terrains constrain the river in a resistant foundation where sediment sources are largely confined to a veneer of glacial deposits. Glacial tills, drift and outwash sands blanket the entire drainage area. Most valleys of the tributaries are lined with unconsolidated silts and clays originally deposited in glacial lakes during the retreat of the Wisconsin glaciation. Ground morain tends to be relatively resistant to erosion, but can supply a wide range of grain sizes to the river. Sand enters the system from local concentrations of glacial sand bodies while silt and clay can be provided from the reworking of glaciolacustrine deposits.

Watershed Sediment Inputs

A valuable synthesis of current understanding of the sediment dynamics in the Hudson River estuary was published by Geyer (2005) following a workshop held at the Hudson River Foundation in November of 2004. Unless otherwise noted, the following text was extracted from that report as it was deemed particularly relevant to understanding the ecological links between the estuary and its watershed.

As part of the NY-NJ Harbor Estuary Program CARP (Contaminant Assessment Reduction Project), Kevin Farley assembled an inventory of inputs to the estuary-harbor system, based on direct measurements where available and using a Normalized Sediment Load approach for other tributaries with more limited data. These estimates indicate an average 860,000 (roughly 1 million) metric tons of sediment are delivered from the watershed (including inputs to Newark Bay). The year-to-year variability of this input is more than a factor of two, with dry years like 1994-1995 showing greatly reduced inputs. In addition, the relative inputs from different watersheds vary considerably from year to year. For example, in 1998-1999, more sediment entered from the New Jersey tributaries than the entire Hudson watershed, whereas in the other years it was only 20-30% of the Hudson total.

Approximately 60% of sediment entering the estuary-harbor system comes from the Hudson-Mohawk drainage, 20% from the New Jersey watersheds (with most going into Raritan Bay), and the remaining 20% from urban drainage.

Average annual sediment loads in thousands of metric tons:

212, Upper Hudson and Mohawk

224, Mid-Hudson, other NY tribs

(436 Hudson-Mohawk Total)

210, New Jersey harbor tribs

71, STP's

83, CSO's

59, Stormwater

(860 Total)

Range during study period was a minimum 468 total in 1994-1995 and a maximum 1185 total in 1988-89.

In the lower reaches of the estuary, watershed sediment sources are supplemented by a supply of sediment up-estuary from the sea (Bokuniewicz 2005). Sand from the Coastal Plain can migrate into the river under tidal influences and fine-grained, marine sediments are recycled into the Hudson by a characteristic, estuarine circulation.

The following was reported in Litten (2005) and was not taken from Geyer (2005). There are few quantitative data on how activities in the watershed might be affecting sediment delivery to the estuary. Litten (2005) makes some speculation about factors driving sedimentation in Hudson River estuary tributaries and sediment export to the estuary. Land use is driven by social and economic factors, and in the Hudson estuary watershed, increasingly by population dispersion. Tributary subwatersheds draining into or near the urban areas of New York City and Albany have high percentages of urban, road, barren, golf course, park, lawn, suburban residential, or cropland land use immediately adjacent to stream channels. Farm land in the Hudson watershed is concentrated in the Schoharie, Mohawk, and Wallkill tributary watersheds. Badly managed farmland can contribute significantly to soil erosion, however well-managed farmland has many environmental benefits. As a number of factors reduce the competitiveness of New York farmland, landowners can become advocates for sprawl type development as a "rational" use for their land. Modern practices, such as extensive suburban development and construction of impervious surfaces, are likely contributors to soil erosion. Over 2,000 dams have been constructed in the entire Hudson River watershed (including non-estuarine portions), and highly regulate stream discharge. The pooled water behind dams promotes sedimentation and can starve downstream reaches of sediment.

A long record of data exists for sediment concentration and load at Waterford. Much shorter records of suspended sediment loads exist from the mouth of the Mohawk River at Cohoes and there are estimated sediment concentrations and loads derived from electronic remote sensing in the Hudson River near Poughkeepsie (Litten 2005). Only four large tributaries are monitored with USGS gage stations in the estuary watershed: the Esopus Creek, Rondout Creek, Wallkill River, and Wappinger Creek. An acoustic Doppler current profiler (ADCP) is now located in the main stem of the Hudson just south of Poughkeepsie. Schoharie Creek has been thought to be an important sediment loading source to the Mohawk River, and perhaps to the Hudson as well. Although this might have been true iin the 1970s, data collected by Litten (2005) suggest the Schoharie Creek behaves similarly to other streams in the Mohawk River watershed. Litten (2005) also reports volunteer sampling in the upper Wallkill has shown that the hypothesis of upland development as a source of sediment to the lower Wallkill is unlikely. Instead, in-stream erosion is probably the major source of sediment to both the Wallkill and the Schoharie.

Litten (2005) recommends a robust monitoring scheme to track the effects of land use changes on sediment delivery to the Hudson estuary. The scheme includes maintaining current monitoring stations at Waterford, Cohoes and Poughkeepsie enhanced by adding stations on the Normans Kill (in Albany), Catskill Creek (below the Kaaterskill in Catskill), Esopus Creek (Mt. Marion), lower Rondout Creek (below the junction with the Wallkill), Roeliff Jansen Kill (at Livingston), and Kinderhook Creek (at Rossman).

Fate and Transport of Sediment in the Estuary

The spring freshet enters the tidal river with high sediment concentrations; most of that sediment settles and is resuspended multiple times in its trajectory through the tidal river. Bokuniewicz (2005) describe two important estuarine sediment transport features: the reversing tidal current that transports sand near the estuary floor up and down the river; and circulation of salty water where suspended sediment is redistributed into turbidity maxima (a region in which the concentration of suspended sediment decreases both upstream and downstream). Fine-grained sediment is transported many times before it is permanently buried in sediment deposits or exported to the sea. It may ultimately be deposited in wetlands, in dredged channels or in undredged areas of the estuary floors.

Kenna and Chillrud (Lamont-Doherty Earth Observatory) used plutonium isotopes from a nuclear processing plant on the Mohawk River to track sediment delivery from the Mohawk. They used chronology of cores to determine that the sediment is initially transported to the lower Hudson (mile 43) in timescales of 0-4 years, but its residence time is on the order of 20 years. This is a surprising result, suggesting that annual throughput of sediment is a small fraction of the sediment that is actually remobilized. It suggests that there are large pools of mobile sediment in the freshwater Hudson that represent as much as 20 times the annual delivery, or approximately 10 million tons. In order to assess the rate at which sediment moves through the river, Wall (United State Geological Survey) compared the sediment flux at the mouth of the Mohawk at Poughkeepsie (mile 72) during the spring freshet in 2003. He found that within 20-25 days, the flux at Poughkeepsie could account for the influx of sediment associated with the spring freshet from the Mohawk. Also in a different study (Lodge 1997), new material was found to have a residence time of 22 days in the estuary (Bokuniewicz 2005).

The lower 18 kilometers of the estuary has been extensively dredged. Suskowski estimates approximately 1.2 million metric tons of annual sediment removal from the harbor for the period 1977-1991.

Klingbeil and Sommerfield (2005) have looked at the sediment accumulation rate in the estuary at a range of time scales. The tidal accumulation rates are extremely rapid, on the order of cm/day. This rapid deposition is highly intermittent and subject to erosional episodes, yielding a net seasonal accumulation rate of approximately 10 cm/year. The net accumulation rate on 100-year timescales is on the order of 1 cm/year, or approximately one tenth of the annual rate. The 1000-year rate of sediment accumulation is about 2 mm/year, close to the rate of sea-level rise. Klingbeil and Sommerfield attribute the systemic decrease in accumulation rate at longer timescales to the occurrence of infrequent erosion events. Events occurring one to two times per century may wipe out 50 years worth of sediment accumulation. Larger events with recurrence intervals of hundreds of years may have even greater erosion potential, accounting for the difference between the 100-year and 1000-year rates of sediment accumulation.

Currently, inputs of sediment to the Hudson estuary from erosion in the watershed are 10-fold higher than if the basin were entirely forested (Swaney et al. 1996). Peteet's research in the marshes of the Hudson estuary show that the rate and character of the sedimentation in the marshes has changed markedly over time--most notably there is a five-fold increase in organic sedimentation since European impact.

Erosion and Deposition Patterns

Nitsche and colleagues have mapped the bed conditions of the entire tidal portion of the Hudson River over the last several years, using side-scan and chirp sonar (Nitsche et al. 2004, Bell et al. 2004). They found that the patterns of erosion and deposition are very heterogenous throughout the tidal river, and every reach of the river has both erosional and depositional portions. These observations provide support for the idea that there is a significant volume of reworking of sediment in addition to the throughput of watershed sources.

An important finding of Klingbeil and Sommerfield (2005) is that large portions of the channel have experienced 1-2 m of erosion in the last 100 years. Analysis of historical bathymetric data indicates that the channel lost an estimated 3 ,000,000 tons of sediment between ca. 1939 and 2002 (50,000 tons/yr average) by subaqueous erosion, increasing in depth by as much as 4 m in places. They attribute these changes to the influence of land fill and pier construction on both sides of the river, which reduced its cross-sectional area by 10-20% in some places. A reduction in cross-sectional area requires a commensurate increase in tidal velocity and a quadratic increase in stress. This increase in stress is hypothesized to have resulted in scour, resulting in a new equilibrium condition with a deeper channel. The state of this adjustment is unknown since the erosion rate cannot be determined within the 60-yr period of the bathymetric change analysis. Also unknown is the local influence of human intervention elsewhere in the system, as intertidal areas throughout the lower estuary and New York Harbor have been reclaimed and the shorelines bulkheaded.

McHugh et al. (2004) used radio-isotopes and high-resolution side-scan sonar surveys to identify regions of sediment trapping in the upper estuary and tidal river between Peekskill (mile 56) and Poughkeepsie (mile 72). They estimate the total trapping in the upper estuary to be approximately 40,000 tons/year. This is roughly 10% of the estimated annual discharge from the watershed, indicating that most of the sediment bypasses the upper estuary and is trapped in the lower estuary deposits. Woodruff et al. (2001) estimate the mass of sediment trapped in the lower estuary during the studies in 1998 and 1999 was approximately 300,000 metric tons per year. This is slightly less but comparable to the estimated sediment flux being delivered from the Hudson-Mohawk, as discussed above. They identified two primary trapping zones in the lower estuary, one centered at river km 13 to the south of the George Washington Bridge, and the other near river km 23 to the north of the bridge.

Bill Ryan and colleagues have performed side-scan and sub-bottom surveys of sediment deposits in the region of the estuarine turbidity maximum zone (Nitsche et al. 2005?). They suggest that the spatial patterns of deposition are controlled not only by the estuarine processes, but also by the accommodation space afforded by historical dredging of the harbor. The dredging activities typically took place in regions of high sediment trapping rates. The increased depth associated with dredging tends to alter the hydrodynamics, favoring sediment trapping. Bokuniewicz (2005) estimates the deposition rates in dredged areas may average 14 cm/yr, although the deposition rate is not uniform. Ryan's point about accommodation space may provide the key to the long-term (decadal and longer) preservation of massive sediment deposits in the region to the south of the George Washington Bridge.

Seabed survey data, radionuclide data and previous studies indicate that another zone of enhanced deposition is located in Haverstraw Bay. Further detailed investigation is needed to confirm the presence of another active ETM.

Geyer et al. (2001) quantified the transport of sediment into and out of the lower Hudson estuary using moored instruments that were deployed between March and June, 1999. This was a relatively dry period, and the spring freshet peak was smaller than average. The observations indicated significant sediment trapping in the lower estuary, as expected. What was not expected was a net northward transport of sediment at both the northern and southern stations, averaged over the deployment period. The authors suggest that fine sediment delivered from the Hudson/Mohawk watershed during pronounced outflow events in 1998 was the source of the sediment transported northward. The relatively weak river flow of 1999 favored the landward transport by the estuarine circulation. Another factor may be the timing of the freshet relative to the spring-neap cycle. If the freshet occurs during the neaps, as it did in 1999, then less seaward sediment transport would be expected.

Knowledge Gaps

(1) Without adequate spatial and temporal observation, the integrated behavior of the estuary remains elusive; (2) Monitoring has not been continuous (storm events are hostile to any measurement program and easy to miss) and most tributaries are not monitored, forcing reliance on indirect calculations of sediment loads; (3) More investigations of the mechanisms and volumes of sediment transport in the tidal freshwater Hudson are needed in order to resolve the contrast in timescales between the physical transport observations (Wall; Lodge 1997) and the geochemical data (Kenna and Chillrud); (4) Little is known of the influence of land fill and pier construction on both sides of the river on channel stability and the status of resulting adjustments in channel cross-sectional area; (5) Further detailed investigation is needed to confirm the presence of another active ETM in Haverstraw Bay.

Cited References

Bell, R.E., R.D. Flood, S.M. Carbotte, W.B.F. Ryan, F.O. Nitsche, S. Chillrud, R. Arko, V. Ferrini, A. Slagle, C. Bertinato, and M. Turrin. 2004. Hudson River Estuary Program Benthic Mapping Project, New York Department of Environmental Conservation Phase II-Final Report, Lamont-Doherty Earth Observatory.

Geyer, W.R. 2005. Final Report Sediment Transport, Erosion and Accumulation in the Hudson River Estuary and Adjacent Water Bodies. A synthesis of a workshop on sediment transport, Thursday, November 20, 2003. Hudson River Foundation, New York, NY. Woods Hole Oceanographic Institution, Woods Hole, MA, March 9, 2005.

Geyer, W.R., J.D. Woodruff, and P. Traykovski. 2001. Sediment transport and trapping in the Hudson River estuary. Estuaries, 24(5): 670-679.

Klingbeil, A.D. and C.K. Sommerfield. 2005. Latest Halocene evolution and human disturbance of a channel segment in the Hudson River Estuary. Marine Geology, 218 (2005): 135– 153.

Litten, S. 200 (draft). Sediments in the Hudson. Report to the NYSDEC Hudson River Estuary Program.

McHugh, C.M.G., S.F. Pekar, N. Christie-Blick, W.B.F. Ryan, S. Carbotte and R. Bell. 2004. Spatial variations in a condensed interval between estuarine and open marine settings: Holocene Hudson River Estuary and adjacent continental shelf. Geology, 32: 169-172.

Nitsche, F.O., R.E. Bell, S.M. Carbotte, W.B.F. Ryan, R.D. Flood, V. Ferrini, A. Slagle, C.M.G. McHugh, S. Chillrud, T. Kenna, D.L. Strayer, and R.M. Cerrato. 2005. Integrative acoustic mapping reveals Hudson River sediment processes and habitats. EOS, Transactions, American Geophysical Union, 86(24): 225-229.

Nitsche, F.O., R. Bell, S.M. Carbotte, W.B.F. Ryan, and R. Flood. 2004. Process-related classification of acoustic data from the Hudson River estuary. Marine Geology, 209: 131-145.

Woodruff, J.D., W.R. Geyer, C.K. Sommerfield, and N.W. Driscoll. 2001. Seasonal variation of sediment deposition in the Hudson River estuary. Marine Geology, 179: 105-119.

Table Of Contents